<br><br><div class="gmail_quote">On Thu, Nov 18, 2010 at 2:08 AM, Rick Kint <span dir="ltr">&lt;<a href="mailto:rwkint@yahoo.com">rwkint@yahoo.com</a>&gt;</span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex;">
<table cellspacing="0" cellpadding="0" border="0"><tbody><tr><td valign="top" style="font:inherit">Hello --<br><br>We&#39;ve experienced a weird problem 3 times in the last month on different clusters.<br><br>Environment:<br>
- Pacemaker 1.0.9.1-1<br>- Corosync 1.2.5-1<br>- RHEL5.4<br>- 2 nodes<br><br>We run a simple configuration with a master/slave resource, appended below.<br><br>Three times we have found different clusters in a state where both instances are slaves and Pacemaker seems perfectly happy with this.  Looking at the configuration (cib.xml) reveals that the target-role attribute for the master/slave resource has changed from Master to Started.  Changing the value of this attribute back to Master fixes the problem immediately.<br>
<br>Any ideas what might be causing this?  Nobody is doing it manually.<br></td></tr></tbody></table></blockquote><div><br></div><div>Sorry to say this, but they are.</div><div>No part of the software manually changes target-role.</div>
<div><br></div><div>Options are:</div><div>1) human</div><div>2) cron job</div><div>3) resource agent</div><div> </div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex;">
<table cellspacing="0" cellpadding="0" border="0"><tbody><tr><td valign="top" style="font:inherit"><br>I&#39;ll hold off on logs until asked, this stuff can get voluminous.<br><br><br>Thanks, Rick
 Kint<br><br><br>&lt;configuration&gt;<br>  &lt;crm_config&gt;<br>    &lt;cluster_property_set id=&quot;cib-bootstrap-options&quot;&gt;<br>      &lt;nvpair id=&quot;cib-bootstrap-options-stonith-enabled&quot; name=&quot;stonith-enabled&quot; value=&quot;false&quot;/&gt;<br>
      &lt;nvpair id=&quot;cib-bootstrap-options-no-quorum-policy&quot; name=&quot;no-quorum-policy&quot; value=&quot;ignore&quot;/&gt;<br>      &lt;nvpair id=&quot;cib-bootstrap-options-dc-version&quot; name=&quot;dc-version&quot; value=&quot;1.0.9-89bd754939df5150de7cd76835f98fe90851b677&quot;/&gt;<br>
      &lt;nvpair id=&quot;cib-bootstrap-options-cluster-infrastructure&quot; name=&quot;cluster-infrastructure&quot; value=&quot;openais&quot;/&gt;<br>      &lt;nvpair id=&quot;cib-bootstrap-options-expected-quorum-votes&quot; name=&quot;expected-quorum-votes&quot; value=&quot;2&quot;/&gt;<br>
      &lt;nvpair id=&quot;cib-bootstrap-options-last-lrm-refresh&quot; name=&quot;last-lrm-refresh&quot;
 value=&quot;1289857566&quot;/&gt;<br>    &lt;/cluster_property_set&gt;<br>  &lt;/crm_config&gt;<br>  &lt;nodes&gt;<br>    &lt;node id=&quot;fnord004&quot; type=&quot;normal&quot; uname=&quot;fnord004&quot;/&gt;<br>    &lt;node id=&quot;fnord005&quot; type=&quot;normal&quot; uname=&quot;fnord005&quot;/&gt;<br>
  &lt;/nodes&gt;<br>  &lt;resources&gt;<br>    &lt;master id=&quot;Bar&quot;&gt;<br>      &lt;meta_attributes id=&quot;Bar-meta_attributes&quot;&gt;<br>        &lt;nvpair id=&quot;Bar-meta_attributes-clone-max&quot; name=&quot;clone-max&quot; value=&quot;2&quot;/&gt;<br>
        &lt;nvpair id=&quot;Bar-meta_attributes-clone-node-max&quot; name=&quot;clone-node-max&quot; value=&quot;1&quot;/&gt;<br>        &lt;nvpair id=&quot;Bar-meta_attributes-master-max&quot; name=&quot;master-max&quot; value=&quot;1&quot;/&gt;<br>
        &lt;nvpair id=&quot;Bar-meta_attributes-notify&quot; name=&quot;notify&quot;
 value=&quot;false&quot;/&gt;<br>        &lt;nvpair id=&quot;Bar-meta_attributes-target-role&quot; name=&quot;target-role&quot; value=&quot;Master&quot;/&gt;<br>        &lt;nvpair id=&quot;Bar-meta_attributes-is-managed&quot; name=&quot;is-managed&quot; value=&quot;true&quot;/&gt;<br>
      &lt;/meta_attributes&gt;<br>      &lt;primitive class=&quot;ocf&quot; id=&quot;Foo&quot; provider=&quot;fnord&quot; type=&quot;foo&quot;&gt;<br>        &lt;operations&gt;<br>          &lt;op id=&quot;Foo-start-0s&quot; interval=&quot;0s&quot; name=&quot;start&quot; timeout=&quot;300s&quot;/&gt;<br>
          &lt;op id=&quot;Foo-monitor-2s&quot; interval=&quot;2s&quot; name=&quot;monitor&quot; role=&quot;Master&quot; timeout=&quot;20s&quot;/&gt;<br>          &lt;op id=&quot;Foo-monitor-3s&quot; interval=&quot;3s&quot; name=&quot;monitor&quot; role=&quot;Slave&quot;
 timeout=&quot;20s&quot;/&gt;<br>        &lt;/operations&gt;<br>      &lt;/primitive&gt;<br>    &lt;/master&gt;<br>  &lt;/resources&gt;<br>  &lt;constraints/&gt;<br>  &lt;op_defaults/&gt;<br>  &lt;rsc_defaults&gt;<br>    &lt;meta_attributes id=&quot;rsc-options&quot;&gt;<br>
      &lt;nvpair id=&quot;rsc-options-resource-stickiness&quot; name=&quot;resource-stickiness&quot; value=&quot;100&quot;/&gt;<br>    &lt;/meta_attributes&gt;<br>  &lt;/rsc_defaults&gt;<br>&lt;/configuration&gt;<br><br><br>
<br></td></tr></tbody></table><br>

      <br>_______________________________________________<br>
Pacemaker mailing list: <a href="mailto:Pacemaker@oss.clusterlabs.org">Pacemaker@oss.clusterlabs.org</a><br>
<a href="http://oss.clusterlabs.org/mailman/listinfo/pacemaker" target="_blank">http://oss.clusterlabs.org/mailman/listinfo/pacemaker</a><br>
<br>
Project Home: <a href="http://www.clusterlabs.org" target="_blank">http://www.clusterlabs.org</a><br>
Getting started: <a href="http://www.clusterlabs.org/doc/Cluster_from_Scratch.pdf" target="_blank">http://www.clusterlabs.org/doc/Cluster_from_Scratch.pdf</a><br>
Bugs: <a href="http://developerbugs.linux-foundation.org/enter_bug.cgi?product=Pacemaker" target="_blank">http://developerbugs.linux-foundation.org/enter_bug.cgi?product=Pacemaker</a><br>
<br></blockquote></div><br>